Académique Documents
Professionnel Documents
Culture Documents
Investment:
Evidence from Non Financial Pakistani Firms
Abstract
The study has been carried out with the purpose to analyze the relationship
between financial leverage and firms investment in the presence of certain
control variables; such as, Tobins Q, cash flow, liquidity, return on equity and
sale etc. The relationship was analyzed by implying different methodologies such
as pooled regression, fixed effect model and random effect model. Housman test
was performed for selection between fixed and random effect model. Data was
taken from balance sheet analysis of joint stock companies, annual reports of the
companies, Business Recorder and Karachi Stock Exchange. Data was collected
for nine years from 2000 to20008, but analyses were performed for eight years
because 2000 was taken as a lag. This study found that financial leverage has
significant negative impact on firms investment.. The results of the common
effect model support that capital structure plays a vital role in the decisions of
firms that is how to invest. But whenever we extended the model to incorporate
the time and individual effect, then no relationship was seen. The relationship
between liquidity and investment is positive but insignificant. Tobins Q has also
shown positive but slightly insignificant relationship with investment for the
target samples. The relationship between cash flow and investment is negative
and is highly significant. Finally the results or ROE indicates that profitability and
investment moves in the same direction.
Introduction
Corporate sector is considered to be the backbone of an economy. Corporations
play a vitalrole in contributing to the economic growth. In todays dynamic
environment, firms generallyface intense competition and should therefore need
to act in response. They make hugeinvestment in modern technology,
infrastructure, land, building, machinery, quality management, innovation and
product development etc. Such factors will help organization to promote
efficiency and effectiveness and gain competitive advantage. A firm needs cash or
money to invest in land, building, machinery and to take care of day to day
operations. The money which firms or businesses invest in purchasing these assets
is called capital investment. In other words we may say that investment is made
on capital goods by firms which ultimately amplified production of consumer
goods.According to Brown (2006), the investment is current commitment of funds
for a period of time in order to derive future payments that will compensate the
investor for the time the funds are committed, the expected rate of inflation and
the uncertainty of future payments. But a very important question rises here that
how such funds will be generated. Literatureture points out mainly two sources of
fund generation, these sources include equity financing and debt financing. Equity
financing means raising funds for company activities or operations by issuing
stocks to individual and institutional investors. These stocks can be common or
preferred in nature. These individual and institutional investors become creditors
and receive ownership interest in exchange for their funds. On the other hand
when a company raises funds through the issuance of bonds or borrowings from
banks or other financial institution, it is called debt financing. Inreturn these
individuals and institutions receive promise that they will receive interest
periodically and the principal amount at maturity. When a company uses mix of
these two sources it is called capital structure. The purpose of this study is to
investigate that whether the financing which a company or firm generates through
debt, effect its investment or not. There is abundant literature regarding the
inclusion of leverage in capital structure of a firm. According to Odit and Chittoo
(2008), in 1930s and 1940s the inclusion of debt in capital structure was
considered as evil as also as taboo and the basic source of bankruptcy and
financial distress. But the concepts were changed by Modigliani and Miller
(1958). They claimed that the value of a firm is independent of its capital structure
in a frictionless world or complete world, where there is no transaction cost, no
default risk, asymmetric information and no taxes. But whenever any assumption
is relaxed the irrelevance value of firm and its capital structure does
not hold any more. If the irrelevance theory is correct then the market value of
firm can increases, as it takes more and more debt. It suggests that high levered
firm will provide tax shield advantage, which in turn increases the value of firm.
Anyhow it is not common practice to observe firms that use only debt for
financial considerations. A question arises here that what are the reasons that
prevent firms from debt financing only. Literature suggests bankruptcy cost and
agency cost to be the primary sources that leads to optimal capital structure.
Therefore agency cost is considered as important issue in corporate sectors. The
separation of ownership and control in corporate sector is a source of agency
problem. It may be in the form of inefficient efforts from manager, investing on
their own preferences, or failing in maximizing shareholder wealth or value of the
firm Bankruptcy costs also discourage borrowing or limit leverage in capital
structure.
Bankruptcy
cost
represents
the
cost
associated
with
liquidation
and
reorganization. Robicheck
and Myers (1966) argue that the cost of financial distress is incurred when the
firm comes under
the threat of bankruptcy, even if it can be avoided.
Theory suggests that the choice of capital structure may help to reduce these
agency cost. Harris
and Raviv (1991) argued that greater leverage may affect managers and it may
reduce agency
cost through the threat of liquidation, which causes personal losses to managers in
the form of
low salaries reputation etc. On the other hand Berger et al. (2006) comments that
when leverage
increases expected cost of financial distress or bankruptcy, the agency cost of
outside debt may
magnify the agency cost of outside equity, so further increase in leverage results
in higher total
agency cost. Therefore they argued that high leverage may cause agency problem
between
bondholder and equity holder. The study of Cantor (1990) highlights the
relationship between
leverage and investment. According to him a firm with a huge amount of average
cash flow can
accumulate a large amount as a reserve which can be drawn upon to make an
investment when a
firm faces shortage of cash flow in a particular year. On the other hand a highly
levered firm
with a small amount of average cash flow does not maintain reserve and may need
to cut
investment back in the year when firm faces shortage of funds. It means that
investment is more
sensitive to cash earning in highly levered firm. Therefore, the leveraged firm
shows greater
variability in its investment over time.
In the past, many empirical studies have been conducted to investigate the
relationship between
leverage and capital investment theories. Lang et al. (1996) concluded with
negative relationship
between leverage and investment, but the intensity was stronger for firms with
low Tobins Q or
low growth opportunities. These findings were consistent with the overinvestment
theory of
(Stulz, 1990). These studies were undertaken in mature markets and developed
economies such
as U.S. and Canada. In these countries long term finance is provided by profit
making
organizations such as banks and other public bonds markets. Here, the debtholders keep an eye
on and control the firms they lend to. However, the results from these studies do
not necessarily
generalize to transitional economies where the relations between lenders and
borrowers are more
complex and subtle. Transitional economies are characterized as having nascent
stock markets,
an absence of public debt markets (or, at most, embryonic public debt markets),
and a reliance on
bank borrowing. Furthermore, most banks are state owned and their decision
making often
reflects the policies dictated by government. Sapienza (2004), Khwaja and Mian
(2005), and La
Porta et al. (2002) argue that state-owned banks are controlled by politicians who
use the banks
to maximize their own political and personal objectives such as providing jobs for
political
supporters and bailing out poorly performing firms. In these cases, the banks'
incentives to exert
disciplinary pressures on firms are compromised. This could be particularly true
for the poorly
performing and loss-making firms because the state-owned banks often have the
obligation to
support them (Cull and Xu, 2003; Dobson and Kashyap, 2006). An investigation
into the link
between leverage and investment in an environment where banks are state owned
can therefore
provide a useful addition to the literature. Privately owned banks in developed
countries
generally use commercial criteria in making lending decisions although on some
occasions
of the benefits from the investment may accrue to debt-holders; this is known as
the debt
overhang problem and leads to underinvestment. Alternatively, Jensen (1986)
argues that high
leverage or (high portion of debt in capital structure) in low growth firms are used
to discourage
management from investing in non-profitable businesses or projects. Here, debt
pre-commits
firms to pay cash as interest and principal and such commitments in low growth
firms can reduce
managerial discretion over free cash flows that may have otherwise been allocated
to negative
NPV projects. In other words, the banks and other debt-holders perform a
beneficial monitoring
and disciplinary role in low growth firms where a high level of debt can limit the
overinvestment
bias caused by managerial agency problems. In recent years empirical studies
have been
undertaken to examine the relevance of the leverage and capital investment
theories. Lang et al.
(1996), Aivazian et al. (2005a), and Ahn et al. (2006) all report a negative relation
between
investment and leverage although the correlation is much stronger for firms with
low growth.
This evidence is consistent with the overinvestment story (Stulz, 1990) where
leverage inhibits
managers of low growth firms from investing in non-profitable capital
expenditures. These
studies use data from the U.S. and Canada where long-term debt finance is
provided by profit
maximizing banks and public bond markets. Here, the debt-holders keep an eye
on and discipline
the firms they lend to. Nonetheless, the results from these studies do not
necessarily generalize to
transitional economies where the relations between lenders and borrowers are
more complex and
subtle. Transitional economies are characterized as having nascent stock markets,
an absence of
public debt markets (or, at most, embryonic public debt markets), and a reliance
on bank
borrowing. Furthermore, most banks are state owned and their decision making
often reflects the
policies dictated by government. Khwaja and Mian (2005), and La Porta et al.
(2002) argue that
state-owned banks are controlled by politicians who use the banks to maximize
their own
political and personal objectives such as providing jobs for political supporters
and bailing out
poorly performing firms. In these cases, the banks' incentives to put forth
disciplinary pressures
on firms are compromised. This could be particularly true for the poorly
performing and lossmaking firms because the state-owned banks often have the obligation to support
Dobson and
Kashyap (2006). An investigation into the link between leverage and investment
in an
environment where banks are state owned can therefore provide a useful addition
to the
literature. Privately owned banks in developed countries generally use
commercial criteria in
contains methodology of the study along with the econometric tests used to find
the relationship.
Section 3 highlights the results and discussions, while section 4 concludes the
study with
findings and applications.
Literature Review
The relationship between capital structure decisions, firm value and performance
has been
studied widely in the last few decades. The finance literature on capital structure
started after the
paper of Modigliani and Miller in 1958. Since then, many capital structure
theories have been
developed to determine the factors that influence capital budgeting decisions such
as; trade -off
theory, pecking order theory, and agency cost model and many more. Modigliani
and Miller
were the pioneers and they studied the relationship between the cost of capital,
corporation
finance and theory of investment. Modigliani and Miller (1958) found in their
paper that in a
perfect world (no transaction cost, bankruptcy cost and no taxes) value of the firm
remained
same whether it is financed with debt or equity. The researchers considered that in
a perfect
world the difference between equity financing and debt financing reduced to one
terminology,
and they considered equity and debt as perfect substitutes. They argued that
investment policy of
a firm should be based only on those factors that would increase the profitability,
cash flow or
Moreover (Myers and Majluf, (1984) argued that investor generally considers an
equity issue as
a sign of overvaluation. Furthermore, Frank et al (2002) demonstrated in their
study, that high
growth firms generally with more financing needs end up with high debt ratio
due to managers
reluctance to issue equity.
Barclay (2001) on the other hand concluded that firm with more growth
opportunities issue
less debts in their capital structure. According to Abor et al (2009) external
finance is more
expensive for riskier securities due to information asymmetry between mangers
and security
holders. Therefore firm prefer to finance its assets and projects with internal fund
first, then debts
and equity as a last resort. The study of Bolton and Scharfstein (1996) compared
the external
sources of financing for funds generations. Further Researchers argued that firms
must finance
its projects with debt rather then equity. They observe US firms for the period of
41 years from
(1946 to 1987) and found that 85% of external financing were raised through
debts as compared
to 7% through equity.
The validity of pecking order theory was tested by Frank and Goyal (2002) for
publically
traded American firms. Researchers used date for 27 years from 1971-1998. The
result of the
study was contrary to those what is often suggested. This study explored that
internal finance
finances its assets through debts, it exposes to such risk. Such risk arises in a
situation when a
company is not able to generate enough cash flows from its operating, financing
and investing
activities to meet its financial obligations.
Tax incentive
Researchers who worked on capital structure have given different explanations of
the impact of
taxation on the capital structure decisions. Eriotis (2007) found that benefit which
is associated
with debt financing is the fact that interest payments are subtracted in calculating
taxable
income, permitted a tax shield for the firms. The tax shield which arises from
borrowing
permits firms to pay lower taxes than they should, when using mix of capital,
instead of using
only their equity financing. Frydenberg (2004) found that as tax rate increases the
value of firm
tax shield will also increase. Firms generally reduce income by paying interest
expenses on debt
and ultimately reduce their tax liabilities. The tax advantage were compared
between large and
small firms by (Pettit and Singer 1985) they found that small firms are generally
less profitable
and made less use of tax shied than large firms and generally have greater chances
of
bankruptcy. Therefore small firms should use less debt in their capital structure as
compared to
their large counterparts. According to (DeAngelo and Masulis1984) the firms that
pays tax,
generally substitutes debt for equity as a minimum to the level where the
probability of financial
distress and bankruptcy starts to be important.
Bankruptcy incentive
Bankruptcy costs are the costs incurred when the perceived probability that the
firm will default
on financing is greater than zero. Bankruptcy costs may be direct on indirect.
Direct bankruptcy
costs include legal expenses, trustee fee and other administrative costs in the
bankruptcy process
(Abor et al 2009). Robicheck and Myers (1966) argue that the cost of financial
distress is
incurred when the firm comes under the threat of bankruptcy, even if it can be
avoided. Myers
(1977), for example, demonstrates that, with sufficiently high leverage, the firms
share holders
do not want to issue new stock due to debt overhang. Owing to this, most projects
with positive
net present value (NPV) can go unfunded and if a firm raises more new debt then
the chances of
bankruptcy increase
Agency cost model
Agency conflicts represent important issue in corporate world, both in financial
and non financial
industries. Such conflicts start to arise as owner of firm delegates authority to
agents. When
owner delegates authority to mangers it results in a conflicts which induces cost,
which is
generally called agency cost. The purpose over here is to provide a brief literature
on agency cost
from the viewpoint of leverage and investment.
Literature suggests that the choice of capital structure may help to lessen these
costs. Agency
cost arises because of the differences in the interest of the ownership and
management; it may
take the form of preferences for the job perks, shirking and self interested
decisions that reduces
shareholder wealth (Ang et al 2000). According to Jenson (1986) debt in capital
structure is in
efficient means of reducing agency cost which is associated with equity.
Furthermore researcher
state that liabilities in capital structure help in avoiding over investment problem,
because the
payment of interest on debt reduces the cash flows left up to corporate managers
and it ultimately
mitigate agency problem. But liabilities have also potentially negative impact of
causing under
investment problems for companies with growth opportunities. Williams (1987)
finding also
support the theory of Jenson that incorporating more debt in capital structure
reduces agency
cost.
Hart and More (1995) emphasized the role of liabilities in reducing agency
problems
between shareholder and managers. Researcher argued that debt increases
efficiency of firm
because it averts managers from undertaking projects with negative net present
value. On the
other hand debt may also block or miss the profitable business investment
opportunities
Leverage and size
Size is the important characteristic of any firm and is videly studied in the
literature. Many
researchers predict a positive relationship between leverage and firm size. Rajan
and Zingales
(1995) studied such a relationship and concluded with a positive sign. They
demonstrate that
leverage of any firm increases as its size expends. Ang et al (1982) investigated
that large firm
are generally more diversified and less prone to bankruptcy. These arguments
support the idea
that large firms are generally more levered as compared to small firms. Titman
and Wessels
(1998) found negative relationship between size and short term debts, possibly
due to reasonably
high transaction costs small firms face when issuing long-term financial
instruments.
Leverage and risk
The variability in return of a firm is generally considered as a proxy for firm risk.
And risk is
assumed to be negatively related to leverage. No significant relationships were
found between
risk and leverage in the study of (Mohammed Amidu 2007).
Leverage and profitability
Literature identified firm performance and profitability as major determinants of
capital
structure. According to tax trade-off models large and profitable firms employ
more debts due to
high tax burden and low bankruptcy risk (Ooi 1999). On the other hand (Myers
1984) and
pecking order hypothesis reports negative relationship between leverage and
profitability, further
researcher argued that profitable companies often dont rely on external fund.
Instead, they
generally used internal funds, which are available to them in the form of retained
earnings. The
study of (Mohammed Amidu 2007) for bank in Ghana revealed a negative
relationship between
level of short term debts and firm characteristics such as, growth, profitability and
assets
tangibility. In addition (Serrasqueiroa and Rogao 2009) also researched the impact
of listed
Portuguese companies specific determinants on adjustment of actual debt towards
target debt
ratio and they concluded negative results between profitability and leverage.
Leverage and growth
The study of Lang et al (1995) showed that there is negative relationship between
leverage and
growth opportunities for firms with low Tobins Q; on the other hand it is positive
for those
firms whos Tobins Q is greater than 1. The pecking order theory supposes
positive relation
between leverage and growth, while trade-off theory assumes negative relations.
The findings of
(Karadeniz et al 2009) indicate that asset tangibility ant return on assets is
negatively related to
debt ration while growth opportunities, size of the firm have no relation to the
debt ratio.
Investment
Investment plays a key role in explaining constant growth. Investment plays a key
role in
explaining constant growth. Investment is studied widely in the last few decades.
The empirical
relationship between investment and q and has been investigated for more than
twenty years with
varying degrees of success. The most basic empirical study of (Furstenburg 1977)
built directly
on Tobin's (1969) argument that investment is an increasing function of q, and
these studies
simply regressed aggregate investment on q. Around that time, Mussa (1977) also
found that a
version of the q theory can be derived rigorously from a model of investment by a
firm facing
convex costs of adjustment.
Different models for investment measurement are given below. These models
were developed to
provide explanation of firm investment. In the literature different models have
been used for
investment determination, which is briefly explained below.
Investment models
From the literature we can identify four models for investment measurements;
namely,
1) Neoclassical model
2) Sales accelerator model
3) Tobins Q model
4) Euler model
Neoclassical model
Jorgenson (1963) proposed a model for investment measurement; the model
explained that cost
of capital is the major and lonely determinant of corporate investment. The model
is defined as
In the given model I is the investment of a firm, K is the capital stock whereas CF
is the cash
flow of firm. The coefficients 1 and 2 and 3 give cash flow sensitivities for
firm i and is
the error term. The model was criticized on the basis that it does not include any
forward looking
variables Goergen and Renneboog (2000).
Sales accelerator model
Sales accelerator model was proposed by (Abel and Blanchard 1986). The model
emphasized
that investment of a firm is determined through only and only sales of the firms.
The model was
criticized on the basis that it does not include growth potentials and assumes that
investment
grows along with sales of the company.
the investment for firm (i) needed to generate future profitability. The model is
good up to some
extent as it covers the limitation of future growth and expectation, but it is not
without problems
The first problem associated with the model is the estimation and measurements
of Tobins Q.
Usually companies do not report the replacement value of their capital stock in
their final
accounts. Some researchers use the book value of assets instead of the
replacement value of the
assets, but it also bears the problem of estimation of intangible assets.
Second, Tobins q will only include future expectations if the firm is a price taker
in perfectly
competitive industries, if there are constant returns to scale and if the stock market
value
correctly measures the fundamental present value of the firms future net cash
flows (Hayashi
1982). In practice, these conditions may not be fulfilled, e.g. if the stock market
displays
excessive volatility relative to the fundamental value of the companies. Thus, if
cash flow (or
profitability) variables are included in an investment model along with Tobins q,
these cash
flow variables may still be made up of expectations not captured by Tobins q. It
may then be
difficult to disentangle the effect of expectations from the one of liquidity
constraints in the
parameter estimate of the cash flow variable. Chirinko and Schaller (1995) show
that average
Tobins q is flawed as it reflects the average return on a companys total capital
whereas it is the
marginal return on capital that is relevant. Gugler, Mueller and Yurtoglu (1999)
develop a
method to measure marginal Tobins q and test the degree of cash flow sensitivity
to investment
in different Tobins q scenarios to differentiate between cases with asymmetric
information and
agency conflicts.
Euler equation model
The Euler-equation model proposed by Bond and Meghir (1994) is based on the
first-order
conditions of a maximization process. The model deals with the shortcomings and
limitations of
the neoclassical and average Tobins q-models. The level of investment relative to
the capital
stock is a function of discounted expected future investment adjusted for the
impact of the
expected changes in the input prices and net marginal output. The Euler
specification has the
advantage that it controls for the influence of expected future profitability on
investment
spending whilst no explicit measure of expected demand or expected costs is
required as future
unobservable values are approximated by instrumental values. The theoretical
model translates
into the following empirical specification and tests the wedge between retained
earnings and
outside financing.
=
+ 3
i +
Where D stands for the debt of the firm, i and i stand for time specific effects
and fixed
effects respectively, and all the other symbols are as previously defined.
Leverage and investment
The relation between leverage and investment opportunities has been a topic of
interest among
finance scholars for many years. According to Odit and Chittoo (2008) in 1930s
and 1940s the
inclusion of debt in capital structure was considered as evil. And it was considered
as the basic
source of bankruptcy and financial distress. Modigliani and Miller (1958)
proposed a theory
regarding capital structure. They claimed that the value of a firm is independent of
its capital
structure in a world with no taxes, no default risk, no transaction cost and perfect
& frictionless
market. Myers (1977), for example, it demonstrates that, with sufficiently high
leverage, the
firms share holder doesnt want to issue new stock due to debt overhang. Owning
to this most
projects with positive net present value (NPV) can go unfunded.
Stulz (1986) predics a negative relation between leverage and investment. He
demonstrates that
investment is negatively related to firm investment and profitability. Therefore
high profit firms
should have a lower leverage. In the last two decades empirical studies have been
undertaken to
study the relationship between leverage and investment decisions. Lang et al.
(1995) report
negative relationship of leverage and firm investment, but its effect is stronger for
small firms
with low growth. Similarly Aviazion et al., (2005) find that a higher percentage of
long term debt
in total debt significantly reduces the firm investment especially in firms with
high growth
opportunities. In contrast he found no significant relation in debt maturity and
firm investment
for firm with low growth opportunities. Likewise Michael et al (2008)
investigated the
relationship between leverage and investment in china. Their result depicts a
negative
relationship between leverage and investment. The researchers further
investigated that negative
relation is weaker in firm with low growth and poor operating companies while it
is stronger for
firms with high growth opportunities and good operating performance. On the
other hand
McConnel and Servaes (1995 find that for firms with low P/E ratio or low growth
opportunities
the value is positively related to the degree of leverage while for high P/E ratio or
high growth
opportunities leverage is negatively associated with firm value. Such finding is
also supported by
Aggarwal and Kyaw (2006). They reveal that leverage is positively and
significantly related to
firms growth for low firms Q ratios. The researchers further suggest that
leverage is value
creating for firms with low growth opportunities while it is value- reducing for
firms with high
growth opportunities.
Norvaisien et el (2008) investigated the relationship between loan capital,
investment and
growth. They demonstrate that firms debts and agency problem (between
managers, shareholder
and creditors) causes underinvestment or overinvestment which in turn has a
negative impact on
corporate investment, growth and firm value. Similarly the finding of Mehmet
Umutlu (2009)
also shown negative relationship between leverage and investment but only for
low growth
firms, and it was the finding of oneway error components model. However, no
relationship was
found when he extended the model to include time effect in a two way component
model.
Similarly Firth et al (2008) examined the relationship between leverage and
investment under a
stateowned bank lending environment in china. They concluded their study with
three major
findings; first that there is negative relationship between debts and growth, second
that this
negative relationship is weaker for firms with low growth opportunities and bad
operating
performance, while stronger for firms with good operating performance and high
investment
opportunities. Thirdly, this relationship is weaker for firms with greater and high
level of state
shareholding, than the firms with low level of state shareholding. They further
explain that state
owned banks in china impose fewer restrictions on investment expenditure of
firms with low
growth opportunities, which create overinvestment problems in these firms. A
strong negative
found in their study that short term debt is significantly helpful in reducing
agency costs on
underinvestment and overinvestment. However, such decline in agency costs
couldnt cause any
increase in leverage level, as the firm early debt level choice depends on the type
of growth
options in its investment opportunity set
Research Methodology
This chapter contains the measurement of variables and the methods and sources
of data
collection. It also highlights different methodologies that have been used to
determine the
relationship between financial leverage and investment.
Sample and Data collection
This section highlights the methods and sources of data collection and sample size
used in the
study. This study; the relationship between financial leverage and firm
investment, is based on
secondary data. The data for the study are extracted from the annual reports,
Karachi Stock
Exchange, Business Recorder and Balance Sheet Analysis of joint stock
companies for the
period of nine years from 2000 to 2009. Data for the study was collected for nine
years but
analysis was conducted for eight years, the year 2000 was taken as lag year.
Initially the samples
included all non financial companies (435) which are listed on Karachi Stock
Exchange, but after
screening the data, the firms with incomplete data were dropped from the sample.
After
screening only 343 firms remained for Panel data estimation. Financial firms,
such as banks,
insurance companies, leasing companies, mudarrabas etc were excluded from
sample due to
regulation complications and different capital structures. The sample includes 343
firms from 27
different sectors. The sector wised classification are as follows.
SECTORS
NUMBER OF FIRMS
TEXTILE SPINNING
74
TEXTILE WEAVING
10
TEXTILE COMPOSITE
37
WOOLEN
6
SYNTHETIC and RAYON
9
SUGAR and ALLIED INDUSTRIES
25
CEMENT
12
TOBACCO
4
REFINERY
6
POWER GENERATION and DISTRIBUTION
6
OIL and GAS MARKETING COMPANIES
5
36
TOTAL NUMBER OF FIRMS
343
Measurement of Variables
For this study we have taken the model from literature for measuring investment.
The model has
been used by Lang et al, (1996) and Aviazian et al. (2005) which is as follows
In the proposed model the Ii-t is the net investment of the firm i at time t; while
Ki,t- 1 is the lagged
net fixed assets; CFi,t is the cash flow of firm I at time t.; Leveragei,t _1 is the
lagged Tobins Q;
Levi,t-1 is lagged leverage; Salei,t-1 is lagged net sales of firm i; is a constant
effect; t-1 is the
error term.
Description of variables
In the proposed model investment i,t is the ratio of net capital expenditure (capital
expenditure
depreciation ) of firm I at time t to total to the lagged net fixed asset; Levi,t-1 is
the lagged one
time period ratio of total liability to total asset. In the literature other dimension
have also been
used for its calculation such as ratio of long term liability to total asset and ratio of
short term
liability to total asset. We will follow prior literature to control for firm cash
flows, Cash flowi,t
is the ratio of operating cash flow of the firm to the lagged net fixed assets. Other
control
variable such as Tobins Q and Sale is also included in the model. It is a proxy for
growth; its
value greater than 1 represent high growth opportunities and less than one
represent low growth
opportunities.
Tobins Q is lagged one time period. It is the ratio of market value of total assets
of the firm to
book value of the firm. The market value of the firm can be calculated as the
value of the
common stock and the estimated value of the preferred stock. Salei,t is the ratio of
net sale to
lagged one time period fixed assets.
Sales will be measured as sale deflated by net fixed assets.
Investment
An important debate in corporate finance is that whether gearing ratio or leverage
ratio effect
investment policies or not. The purpose of this study is also to check such
relationship. We have
measured investment as the ratio of net investment to lagged fixed assets. Net
investment was
calculated as (capital expenditure depreciation). The same ratio has been used
by (Lang et al.
1995, Aivazian et al. 2003 and Odit and Chittoo. 2008.)
=
Leverage
In this study financial policy of a firm is measured with leverage. There are
different measures of
leverage, such as Long term debt to total assets, short term debt to total assets and
total liability
to total assets. We have used total debt to total assets as a measure of leverage.
The same
measure has been used by (Pamela et al. 1983, Mehmat Umutlu 2009, Ahn et al
2005 and Sean
Cleary. 1999).
Control Variables
constrained firms. Cash flow is the ratio of cash flow before extraordinary income
to lagged
fixed assets.
This ratio has also been used by (Odit and Chittoo 2008).
Tobins Q
Tobins Q measures the performance of a firm and it is the ratio of (market
capitalization of the
firm + book value of the debt) to book value of the assets. From the literature it is
clear that
Tobins Q has a significant impact on investment. Gomes (200I) describes that
investment of a
firm in very much sensitive to Tobins q and Cash flow. Researcher further argued
that the
availability of external financing makes no difference. The idea was also
supported by Cooper
and Ejarque (2003) they solved a model with quadratic adjustment costs and a
concave revenue
function, and also found that investment is strongly related to Tobins q and cash
flow.
In order to find out the exact relationship between financial leverage and firm
investment we
controlled for both Tobins Q and Cash glow.
Sale
Sale of a company is also taken in to consideration, it is calculated as follows.
ROE
To control for the impact of profitability on firm investment, we have taken Roe
as` a proxy for
firm profitability. It is the ratio of net income to shareholder equity.
The Ratio was also used by Odit and Chittoo (2008).
LIQUIDITY
The liquidity ratio is measured by the current assets divided by the current
liabilities.
Panel Data Estimation
This study checked the relationship between financial leverage and firm
investment. The
relationship was checked by applying panel data. From panel data we get two
dimensions of
data, such as time series and cross sectional. Panel data has certain advantages,
such as it offers a
wide range of observations which better estimate the parameters. According to
Dimitrious
Asteriou (2006) panel data outsources more information to the analyst and
accounts for the
dynamic behavior of parameters. Panel data was estimated by three different
methodologies
Pooled Regression or common constant model, fixed effect model and random
effect model
Common Effect Model
Common effect model or pooled regression analyzes the relationship with the
assumption that all
the intercept will remain constant all over the years and across sectors. Practically
the common
constant method implies that there are no differences between the estimated cross
section and it
is useful under the hypothesis that the data set is a priority homogeneous. This
assumption is too
much restrictive and beyond the reality. To take into consideration the time effect
and individual
effect we had applied the fixed effect and random effect model.
Fixed Effect Model
Fixed effect model is similar to polled regression but it allows for the constant to
vary across
individuals. It is also called Least Square Dummy Variable (LSDV) estimator,
because it uses
dummy variables for taking different constants in to account (Gujrati 2006).
Different intercept
concepts are logical because our samples consist of heterogeneous set of non
financial firms
relating to diverse sectors.
i,t = + 1Xi,t + ei,t
i,t = dependent variable of firm i at time t.
= intercept
1 = slope of the independent variable
Xi,t = independent variable of firm i at time t.
ei,t
= error term of firm i at time t.
Dummy variable is the one that allows us to take different group specific
estimates for each of
the constants for every different section. Despite of its strength the model also
faces certain
problems such as;
Firstly it ignores all explanatory variables that dont vary over time.
Secondly it is inefficient because it estimate a very large number of parameters.
Therefore it is
not good to use fixed effect model without considering another model such as
random effect
model.
Random Effect Model
The random effect model does not take intercept as constant or fixed but as
random parameters.
Random effect model assumes that there is not a constant or fixed intercept for
each company
but a random drawing from much larger population with a common mean value
for the intercept.
As this model doest discriminate between the intercept of companies, it allows
the error terms to
take into considerations all the differences in the individual intercept.
i,t = 1 + 2 Xi,t + i + ui,t
i,t = Dependent variable of firm i at time t.
1 = Mean value of all intercept
2 = Slope of the independent variables.
Xi,t = Independent variable of firm i, at rime t.
i = Deviation of individual intercept from mean.
ui,t = Error term of all the firm i, at time t.
One major disadvantage of random effect model is that it needs to make specific
assumptions about the distribution of the random components. Also if the
unobserved group
specific effects are correlated with the explanatory variables then the estimates
will be biased and
inconsistent. Having this disadvantage this model has certain advantages as well.
First, it has fewer parameters to estimates as compared to fixed effect model.
Secondly it
allows us for additional explanatory variables that have equal values for all
observations within a
group.
In general the difference between the two possible ways of testing panel data
model are
that the fixed model assumes that each variable differs in its intercept terms, while
random model
assumes that each variable differs in its error term. According to Dimitios
Asteriou when the
panel is balanced the fixed effect model will work excellent. On the other hand
when the sample
contains limited observations of the existing cross sectional unit the random
effects model might
be more appropriate. But statistically Housman test will show that which model is
appropriate to
use.
Housman specification test
A Housman test is formulated in the selection of fixed effect and random effect.
Housman state
the hypothesis that
H0 = fixed effect model and random effect model estimators are not different.
H1 = fixed effect model and random effect model estimators are different.
Results and Discussion
The purpose of this section is to provide detail empirical evidence of the study.
These empirical
evidences are structured to provide in depth results which include descriptive
statistics,
correlations, three panel data models such as common effect model, random effect
model and
fixed effect model. For comparison between fixed effect and random effect
models the Housman
tests were performed.
Descriptive Statistics Table 1
This table provides summary of descriptive statistics of the dependent and
explanatory variables
used in the study. The sample of the study consists of 342 non financial firms
which were listed
on Karachi Stock Exchange during 2000 2008, with a balance panel of 2736
observations. The
mean of the investment to fixed assets is (-.0420846) while its standard deviation
is (.9877884).
Investment Leverage Liquidity Sale
Cash Flow ROE
Tobin's
Q
Mean
-0.0420846 0.712605 1.726046 9.303822 0.3858371 0.296877 1.155924
Standard Deviation
0.9877884 0.487996 5.839397 54.54453
1.544653 3.217851 1.355087
Minimum
-46.08571 0.018124 0.009915
-47.1111
-44.38889
-41.579 0.078104
Maximum
5.495758 9.117647 265.0227 1236.261
38.05714 137.6667 42.93401
The results show that the firm investment is negative on average. And its standard
deviation is
quite high. This implies that the investment of Pakistanis firms move on either
directions. On the
other hand there is a greater variation in investment with a minimum value of (46.08571) and a
maximum value of (5.495758). The mean value of leverage is (.7126046) with a
standard
deviation of (.4879956). The minimum level of debt that Pakistani firms use is
(.0181237) while
the maximum limit for the studied firms is (9.117647). The mean ratio for
liquidity is
Cash
Flow
ROE
Tobin's Q
Investment
1
Leverage
-0.0106
1
Liquidity
-0.0041
0.0288
1
Sale
-0.0576
-0.0409
-0.0009
1
Cash Flow
-0.05464
-0.0737
0.008
0.1718
1
ROE
-0.0048
-0.0137
-0.0002
-0.0017
0.1486
1
Tobin's Q
0.0309
0.2486
0.0068
0.0403
-0.0275
-0.0344
1
In this study we conducted correlation analysis to check out whether there exists
multi-colinearity in the model or not. As Cuthbertson (1996) pointed out in his book that
multicollinearity
exists in the model when the explanatory variables are strongly related to each
other. In this
study multicollinearity were checked by checking the
correlation between Leverage, Liquidity, Tobins Q, Sale, Cash Flow and Return
on Equity. All
the coefficients are less then (.148), which suggests that multicollinearity is not a
serious
problem in our model. The relationship between leverage and investment is
negative which an
indicator of inverse relationship between these two variables. Investment has a
positive
relationship with Tobins Q and negative relationship with all other explanatory
variables.
Leverage has a positive relationship with liquidity and Tobins Q while it has a
negative
relationship with Sale, Cash Flow and Roe. Liquidity represents a positive
relationship with cash
flow and Tobins Q and negative relationship with all other variables. Tobins Q
has a negative
relationship with ROE and Cash flow while positive relationship with all other
variables. Cash
flow has a also a positive relationship with sale and liquidity.
Common Effect Model
Table 3
Variable Description
Coefficient
z statistics
P value
Intercept
0.1410078
4.41
0
Leverage
-0.1076574
-3.08
0.002
Liquidity
0.0009181
0.34
0.736
Sale
0.0007379
2.39
0.017
Cash Flow
-0.3737831
-35.18
0
ROE
0.0251835
5.1
0
Tobin's Q
0.1410078
1.54
0.124
Wald Chi
1256.04
R Square
0.3433
The value of overall R square is .3093; it means that there is 30.93% variation in
the dependent
variable owing to independents variables. The value of overall R square is not
highly
satisfactory, but it is acceptable for panel data. There may be certain other
variables which also
influence the investment decisions of the firms that is way the value of R square is
.3093. The
results revealed that the level of debt does have a significant negative impact on
firm investment.
The coefficient value of leverage is (-.1076574). It indicate that when leverage of
a firm is
increased by 1 unit its investment decreased by (.1076574) units. These findings
support the
theory that leverage has an important role in over coming over investment. It
supports the inverse
relationship between financial leverage and firm investment. These results are
consistent with
those of the Firth et al (2008), they also concluded inverse relationship between
these two
variables. McConnell and Servaes (1995) also concluded that the value of U.S.
firms is
negatively correlated with leverage for high growth firms (indicated by high
Tobin's Q), and
positively correlated with leverage for low growth firms (or low Tobin's Q). These
results are
also consistent with that of Lang et al (1996) they found that leverage is
negatively associated
with investment but only for firms with low growth opportunities. Aivazian et al
(2005) also
found negative association between leverage and investment, and its effect
stronger for firm with
low growth opportunities as compared to high growth opportunities. Similarly
Ahn et al. (2006)
document that the negative relation between leverage and investment in
diversified firms is
significantly stronger for high Q segments than for low Q business segments, and
is significantly
stronger for non-core segments than for core segments. Among low growth firms,
the positive
relation between leverage and firm value is significantly weaker in diversified
firms than in
focused firms The results of Polled regression indicate that capital structures
plays a very
important role in firm investment decisions. The relationship between liquidity
and investment is
positive but insignificant. Tobins Q has also shown positive but slightly
insignificant
relationship with investment for the target samples.
The regression coefficient of Sale is (.0007) and is significant with P value less
then (.05) and z
value greater than 2. Its result indicate that I unit increase in sale leads to (.0007)
units increase
in investment. It means that both are in the same directions, when sale of a firm
increases its
investment is also increased.
The relationship between cash flow and investment is negative and is highly
significant. 1 unit
increased in cash flow causes (.3737) unit decrease in investment.
Finally the results or ROE indicates that profitability and investment moves in the
same
direction. When profitability increases by 1 unit, investment of Pakistanis firms
increases by
(.025) units. It is significant at (000) level and its t value is greater than 5.
Fixed Effect Model
Table (4)
Variable Description
Coefficient
t statistics
P value
Intercept
0.120401
2.94
0.003
Leverage
-0.0328595
-0.67
0.500
Liquidity
0.0033532
1.09
0.275
Sale
0.0012827
2.87
0.004
Cash Flow
-0.4248858
-35.23
0.000
ROE
0.0285548
5.48
0.000
Tobin's Q
-0.0012129
-0.08
0.936
F statistics
209.63
R Square
0.345
The fixed effect model shows that that there is no significant relationship between
leverage and
investment. It means that a firm financial policy is irrelevant to its investment
decisions. The
results suggest that whenever we extend the model to incorporate the time effect
the relationship
between financial leverage and investment disappears. But the results of Sale,
Cash flow and Roe
is somehow similar to that of common effect model. Results support the recent
research findings
Leverage
-0.1076574
-3.08
0.002
Liquidity
0.0009781
0.34
0.736
Sale
0.0007379
2.39
0.017
Cash Flow
-0.3737831
-35.18
0.000
ROE
0.0251835
5.1
0.000
Tobin's Q
0.0189451
1.54
0.124
Wald Chi
1256.04
R Square
0.3433
The results of the random effect model are mostly similar to that of common
effect model.
ROE
0.0285548
0.0251835
0.0033713
Tobin's Q
-0.002129
0.0189451
-0.0210741
Chi 2 (6)
91.59
Prob > Chi 2
0.000
In order to find out which methodology is appropriate, whether fixed effect or
random effect, we
applied Housman specification test. The random effect model assumes that there
is no correlation
between the group specific random effects and the regressors. However, the fixed
effects model
does not make such assumptions and the possibility remains that the assumption
of zero
correlation in random effects model is not feasible. The Hausman test checks
whether the
correlation assumption is statistically evident or not. The null hypotheses for the
Hausman test is
that the group specific random effects and the regressors are not correlated and
thus if the
Hausman test shows a parameter value of more than 0.05 then it would mean that
fixed effects
model is inefficient and random effects model is better (Girma,2006).
The result of Housman suggests that fixed effect model is appropriate to use.
Housman rejects
the null hypothesis that there is no significance difference in fixed effect and
random effect. So
we accept the alternative that both the models are not the same. On the basis of
Chi square and P
value <.05 we select that fixed effect model is appropriate to use.
On the basis of p values it is decided that which model is appropriate whether
fixed effect model
or random effect model
Discussion
The calculated F value is greater than table value. Thats why the chosen variables
are
significantly related with investment during the period. Further it shows that the
leverage has
negative impact on investment. This implies that as leverages increases,
investment in Pakistani
firms decreases. This means that Pakistani firms have adequate resources
financing the projects.
So Pakistani firms are not dependent on debts as a source of finance to finance the
projects.
As we have seen positive relationship between investment and ROE, it indicates
the operating
efficiency of the employed funs over investment is positive. Positive relationship
also attracting
funds from investors for expansion and growth.
Liquidity is positively related to investment and is not statistically significant. It
means that the
failure of a firm to meet its obligation due to lack of sufficient liquidity will result
in poor credit
worthiness loss of creditors confidence. So Pakistani firms should make certain
that that they do
not suffer from lack of liquidity, as this may lead to financial distress and
ultimately bankruptcy.
The result depicts that when profitability increased by 1 unit, investment of
Pakistanis firms was
increased by (.025) units. It was significant at (000) level and its t value was
greater than 2. Sales
had a positive and significant impact on investment, which shows that as sales of
Pakistani firms
increases its investments also move in the same direction.
The value of overall R square is .3093; it means that there is 30.93% variation in
the dependent
variable owing to independents variables. The value of overall R square is not
highly
satisfactory, but it is acceptable for panel data. There may be certain other
variables which also
influence the investment decisions of the firms that is way the value of R square is
.3093. The
results revealed that the level of debt does have a significant negative impact on
firm investment.
The coefficient value of leverage is (-.1076574). It indicate that when leverage of
a firm is
increased by 1 unit its investment decreased by (.1076574) units. These findings
support the
theory that leverage has an important role in over coming over investment. It
supports the inverse
relationship between financial leverage and firm investment for Pakistani firms.
These results are
consistent with those of the Firth et al (2008), they also concluded inverse
relationship between
these two variables. McConnell and Servaes (1995) also concluded that the value
of U.S. firms is
negatively correlated with leverage for high growth firms (indicated by high
Tobin's Q), and
positively correlated with leverage for low growth firms (or low Tobin's Q). These
results are
also consistent with that of Lang et al (1996) they found that leverage is
negatively associated
with investment but only for firms with low growth opportunities. Aivazian et al
(2005) also
found negative association between leverage and investment, and its effect
stronger for firm with
low growth opportunities as compared to high growth opportunities. Similarly
Ahn et al. (2006)
document that the negative relation between leverage and investment in
diversified firms is
significantly stronger for high Q segments than for low Q business segments, and
is significantly
stronger for non-core segments than for core segments. Among low growth firms,
the positive
relation between leverage and firm value is significantly weaker in diversified
firms than in
focused firms The results of Polled regression indicate that capital structures
plays a very
important role in firm investment decisions. The relationship between liquidity
and investment is
positive but insignificant. Tobins Q has also shown positive but slightly
insignificant
relationship with investment for the target samples
Conclusion
This study examined the relationship between financial leverage and firm
investment, for
support to the agency theories of corporate leverage and especially to the theory
that leverage has
a disciplining role in overcoming the overinvestment problems. Thus the result of
the study is
consistent with the hypothesis that leverage attenuates to invest in poor projects.
The results of
the common effect model support that capital structure plays a vital role in the
decisions of firms
that how to invest. But whenever we extended the model to incorporate the time
and individual
effect, then no relationship were seen. The relationship between leverage and
investment were
checked in the presence of certain control variables. The result shows negative
relationship
between investment and cash flow. Strong relationship was seen between
investment and cash
flow. The results indicate that investment and cash flow sensitivity is very strong
for Pakistani
non financial firms. Whenever we checked the relationship between investment
and ROE (proxy
for profitability) positive relationship were found by applying common effect
model. The result
depicts that when profitability increased by 1 unit, investment of Pakistanis firms
was increased
by (.025) units. It was significant at (000) level and its t value was greater than 2.
Sales had a
positive and significant impact on investment, which shows that as sales of
Pakistani firms
increases its investments also move in the same direction.. No relationship was
seen between
liquidity and investment for the selected sample. Tobins Q had also showed
insignicant
relationship with investment. We may therefore concluded that despite different
debt markets
and banking structures between developed countries and developing countries the
relationship
remained same for financial policy (measured as financial leverage) and
investment.
Limitations of the study
There are various methods for calculating financial leverage but this study
employed only one
method; that is total long term liability under total assets, so future research may
be conducted by
employing different methodologies, such as total liability by total assets and long
term liability
by total assets. This study is only limited to the non financial sector of Pakistan.
Thirdly this
study uses book value of debts for financial leverage.
Recommendations for future research
The empirical model of this study can be extended which could generate further
information.
The model can be extended by various ways; by enlarging panel data set or by
increasing the
number of predictors variables. Increasing the data set will further enhance the
applicability of
the research. Secondly, company classification, whether it is a high growth
company or low
growth company has been entirely excluded from the study so future research
may be conducted
in a way that classifies between companies. Thirdly, future research may also be
made on sector
Harris, M., Raviv, A., (1991). The theory of capital structure. Journal of Finance,
Vol` 46, pp.
297355.
Hayashi, F.,( 1982). Tobin's average q and marginal q: a neoclassical
interpretation.
Econometrica, Vol. 50, pp. 213-224.
Jensen, M.C., (1986). Agency costs of free cash flow, corporate finance, and
takeovers.
American Economic Review 76, 323329.
Jo, H., Pinkerton, J.M. and Sarin, A. (1994), Financing decisions and the
investment
opportunity set: some evidence from Japan, Pacific-Basin Finance Journal, Vol.
2, pp. 227-42.
Lang, L., Ofek, E., Stulz, R.M., (1996). Leverage, investment and firm growth.
Journal of
Financial Economics , Vol 40, pp. 329.
McConnell, J.J., Servaes, H., (1995). Equity ownership and the two faces of
debt. Journal of
Financial Economics, Vol 39, pp. 131157.
Modigliani, F., Miller, M.H., (1958). The cost of capital, corporation finance,
and the theory of
investment. American Economic Review, Vol. 53, pp. 433443.
Modigliani, F., Miller. M.H., (1963). Corporate income taxes and the cost of
capital: a
correction. American Economic Review, Vol. 53, pp. 433-43.
Myers, S., (1984) The capital structure puzzle. The Journal of Finance, Vol 39,
pp. 575-92.
Myers, S., Majluf, N., (1984). Corporate financing and investments decisions
when firms
have information that investors do not have. Journal of Finance Economics. Vol.
13, pp. 187-
222.
Myers, S., (1977). Determinants of corporate borrowing. Journal of Financial
Economics, Vol
5, pp. 147175.
Myers, S.C., (2001). Capital structure. Journal of Economic Perspectives Vol.
15 (2), pp. 81
102.
Nishat, M. and Ullah, W. (2008), Capital structure choice in an emerging market:
evidence
from listed firms in Pakistan, paper presented at the 21st Australasian Finance
and Banking
Conference.
Odit, M. P., and Chittoo, H. B., (2008). Does Financial Leverage Influence
Investment
Decisions? The Case of Mauritian Firms. Journal of Business Case Studies.
Volume 4.
Stulz, R.M., (1990). Managerial discretion and optimal financing policies.
Journal of Financial
Economics, Vol. 26, pp. 327.
Ooi, J. (1999), The determinants of capital structure: evidence on UK property
companies,
Journal of Property Investment and Finance, Vol. 17 No. 5, pp. 464-78.
Pettit, R. and Singer, R. (1985), Small business finance: a research agenda,
Financial
Management, Vol. 14 No. 3, pp. 47-60.
Rajan, R., Zingales, L., (1995). What do we know about capital structure? Some
evidence from
international data. Journal of Finance, Vol. 50, pp. 14211460.
Titman, S. (1984), The effect of capital structure on a firms liquidation
decisions, Journal of
Financial Economics, Vol. 13 No. 1, pp. 137-51.
Titman, S., Wessels, R., (1988). The determinants of capital structure choice.
Journal of
Finance, Vol. 43, pp. 119.
Van Horne, J.C. (2002), Financial Management and Policy, 12th ed., PrenticeHall, Englewood
Cliffs, NJ.
White, H., (1980). A heteroskedasticity-consistent covariance matrix estimator
and a direct test
for heteroskedasticity. Econometrica, Vol. 48, pp. 817838.
Zwiebel, J. (1996), Dynamic capital structure under managerial entrenchment,
The American
Economic Review, Vol. 86, pp. 1197-215